|
Stack Machine posted:Digging this up despite the fact that the conversation has moved on cause dimming of LEDs is actually a thing I'm fairly familiar with. WS2812s and that line of LED chips is actually pretty sophisticated for how cheap they are. They use an on chip programmable constant current supply which is how they handle the dimming. I'm sure there's someone out there that has measured current and luminosity with regards to programmed level. And I'm also sure that it isn't sufficient for analytical uses (although there are ways around this such as using splitters with a reference diode or monitoring the input current). They're definitely a hell of a lot better though than deuterium lamps, though if you need lower wavelengths, UV LEDs are still in the infancy stage for the most common detection wavelengths of 200-230 nm. echi, if you want to know how much current it's supplying, just use your multimeter and put it inline from the output of that power supply. That'll tell you how much current it's actually supplying. You're probably okay putting those in parallel if they're the same kind.
|
#
?
Jan 23, 2022 16:07
|
|
|
|
| # ? Apr 25, 2024 19:48 |
|
|
Woolwich Bagnet posted:WS2812s and that line of LED chips is actually pretty sophisticated for how cheap they are. They use an on chip programmable constant current supply which is how they handle the dimming. The constant-current supply sets the brightness at 100% and the fact that it's well-controlled keeps lot-to-lot brightness variation and variation with supply voltage down. This is still sophisticated compared to the old-fashioned method of just using a resistor, but I'm pretty sure the dimming for the 256 software-programmable levels is done with PWM because it's still cheaper to do that than an 8-bit current-output DAC. I'll check when I get a chance since I have a string of the things on my bench but last time I set it up I could see the flicker so I'd be very surprised if it wasn't using PWM for that. E: if I aim a photocell at the LED it looks like it' PWM at 1.3kHz and R,G, and B all switch on simultaneously.   It looks like the codes are linear out of 255, with 0 being 100% off and 255 being 100% on. Stack Machine fucked around with this message at 18:36 on Jan 23, 2022 |
|
#
?
Jan 23, 2022 16:38
|
|
|
yeah that goes with everything Ive read about them. I’ve read there are ways that you can access them more directly and get 1024 levels of pwm but I don’t know how to do that. Woolwich : oh yeah the multimeter, always forget about that dude.
|
|
#
?
Jan 23, 2022 19:03
|
|
|
echinopsis posted:https://www.tiktok.com/embed/7056361498527943938 this feels like a final fantasy boss fight is about to start and i mean that in the best way possible
|
|
#
?
Jan 23, 2022 21:04
|
|
|
PIZZA.BAT posted:this feels like a final fantasy boss fight is about to start and i mean that in the best way possible excellent analogy yes. i love the light
|
|
#
?
Jan 23, 2022 21:08
|
|
|
PIZZA.BAT posted:this feels like a final fantasy boss fight is about to start and i mean that in the best way possible thanks man, means a lot.
|
|
#
?
Jan 23, 2022 22:48
|
|
|
burned out TWO separate 9-axis IMU modules because i bought them years ago and forgot that they were 3.3v only. 5v on the i2c pins is all it took to fry them. ugh! only 10 bucks apiece but still. mad. i wish all 5v stuff would all just go away so i could quit worrying about it and adding dumb level shifters to my projects
|
|
#
?
Jan 27, 2022 19:51
|
|
|
Also, this is something that I've been wracking my brain over that I am pretty sure would be simple for some of the math geniuses in here. I have a three-axis magnetometer and I only want a compass heading out of it. I can do that easily by taking the atan2 of the X and Y components from the magnetometer, as long as the device remains parallel to the earth. If I tilt the Z axis, the reading is no longer accurate. How can I compensate for that? I know that it's going to be problematic if the device is rotated 90 degrees, so I'm fine limiting it to like +/- 30 degrees from vertical or so. Is it even possible without an accelerometer?
|
|
#
?
Jan 27, 2022 20:24
|
|
|
depends, are you near the north pole?
|
|
#
?
Jan 27, 2022 20:40
|
|
|
You could try compensating by using the Z axis value to come up with a 3D magnitude and then making assumptions about which are your X and Y axises, but who knows. The IMUs I used in the past had terrible magnetometers.
|
|
#
?
Jan 30, 2022 06:51
|
|
|
i have a long-term on/off project of setting up a key system and a few phones so far i've cleaned up and repaired this western electric key system phone and acquired a ksu (line cards and backplane) made by itt recently i found and bought this 24 analog line cisco voip gateway for $30 shipped from ebay  i'd love a real landline, or at least a non-voip one over gpon from the ont, but that's not in the cards at the moment
|
|
#
?
Jan 30, 2022 07:02
|
|
|
could you use two orientated 90 from each other?
|
|
#
?
Jan 30, 2022 07:26
|
|
|
i guess i should have posted this here: PIZZA.BAT posted:paging sagebrush or anyone else who's into 3d printing:
|
|
#
?
Feb 14, 2022 17:59
|
|
|
 giant pitaHow interested are you in doing 3D printing as a hobby? An ender 3 is a decent printer, but it doesn't have many of the nice features that newer/fancier machines do to make printing easier and more reliable. There will always be a little bit of loving around with it from time to time. If you're okay with that, then sure, it's a printer that prints well and you can't beat the price. There are a few small upgrades that the people in the 3D printing thread recommend you make ASAP for quality of life. I would ask your brother-in-law if it's currently working well, though, because if you don't know anything about the field you probably don't want to start with a machine that's already flaky, or worse, someone else's broken project with unknown modifications. Sagebrush fucked around with this message at 18:09 on Feb 14, 2022 |
|
#
?
Feb 14, 2022 18:07
|
|
|
Sagebrush posted:How interested are you in doing 3D printing as a hobby? i'm gonna say 'kind of interested'. i'm working on other projects now but when i wrap them up i could see the availability of a printer steering my next foray into something more physical, which could be neat. in the immediate future it'd definitely be collecting dust, though
|
|
#
?
Feb 14, 2022 18:20
|
|
|
go for it, if it currently worksSagebrush posted:There are a few small upgrades that the people in the 3D printing thread recommend you make ASAP for quality of life. if the extruder arm is plastic, get the metal one because the plastic arm will break sooner or later make sure the XT60 connectors on the wires from the PSU to the main board are soldered, not crimped. or just replace the wires. stiffer bed springs (those yellow ones) are a good and easy upgrade if your brother in law was using it then it might already have some of these (and some versions of the ender3 already have these). there is of course a bunch of other upgrades and stuff that's not necessary, but those 3 are fairly cheap and easy upgrades
|
|
#
?
Feb 14, 2022 21:13
|
|
|
Apparently a thing I get the urge to do every now and then is write implementations of Deadfish in rando programming languages I'm looking at, or playing with (or sometimes actually *using*....). Yesterday the itch struck again and I wrote two more: https://esolangs.org/wiki/Deadfish#Umka https://esolangs.org/wiki/Deadfish#ICI
|
|
#
?
Feb 19, 2022 04:33
|
|
|
sagebrush was it you talking about how young people didn't know how much force to use for things because this is completely a thing and i finally understand it now
|
|
#
?
Feb 20, 2022 06:41
|
|
|
do you mean like screwing poo poo for example? computers and technology today is too compelling and never ending, and must generate a big enough reward that physical poo poo like tinkering is just unappealing. not to everyone of course but just less people
|
|
#
?
Feb 21, 2022 00:15
|
|
|
yes exactly that a new guy at work (where i am also new) just could not figure out why this finishing screw wasn't going in to secure an outlet plate (it was bent) and didn't know how much force to use, so the threads never engaged, and he kept trying for 30 minutes and it continued to not work. a broom handle also came off the bristles and he could not figure out how to screw the handle back on
|
|
#
?
Feb 21, 2022 01:31
|
|
|
jfc between that and the people who don't understand the concept of folders i feel like i am 99% of the way to being that old boomer that thinks everyone around me is incomprehensibly stupid
|
|
#
?
Feb 21, 2022 01:50
|
|
|
like the screw in the parable, some people just need some help getting their threads engaged screws don't exist in the state of nature and everybody who can guide a flathead screwdriver into a slot-head screw around a corner and tighten it adequately by feel alone developed this skill through repeated effort and mistake-making brains treat tools as extensions of ones body but you need training to develop your abilities to this point i know a person born in the late 90s who recently went from computer touching only to taking apart and rebuilding an early 90s toyota over a few months, so acquiring mechanical skills seems to mostly have to do with exposure and training
|
|
#
?
Feb 21, 2022 03:26
|
|
|
a few days ago i watched a grad student snap off what looked like an m3 bolt because she was holding a ratchet by the handle on such a small screw, instead of closer to the shaft or using a screwdriver handle. i made the same kind of mistake once or twice years ago. i think like computers, breaking things is part of learning how mechanical stuff works. the penalty is often shittier, but may also involve learning new skills like how to use a screw extractor and heli-coils.
|
|
#
?
Feb 21, 2022 03:31
|
|
|
Wild EEPROM posted:sagebrush was it you talking about how young people didn't know how much force to use for things because this is completely a thing and i finally understand it now it might have been me mentioning that, yeah. i encounter it a lot with students using impact drivers to put screws into things. they often just rest the driver bit in the screw and pull the trigger all the way, and it just goes BZZZZZZ and strips the poo poo out of the screw head. i end up having to tell some of them "put your whole body weight into it" to get them to understand, even though you probably only need like 20 pounds at most. there are a lot of other tools that they screw up in similar ways, but most of them are at least a little more obscure. e.g. not knowing that files only cut in one direction, so they're just rubbing it all over the surface of their part like sandpaper and wasting a ton of effort. i don't expect people to automatically know how to use a file, but drat, how can you get to be an adult without turning a screwdriver? Sagebrush fucked around with this message at 04:29 on Feb 21, 2022 |
|
#
?
Feb 21, 2022 04:27
|
|
|
Sagebrush posted:there are a lot of other tools that they screw up in similar ways, but most of them are at least a little more obscure. e.g. not knowing that files only cut in one direction, so they're just rubbing it all over the surface of their part like sandpaper and wasting a ton of effort. There's actually a really neat video about this: https://www.youtube.com/watch?v=xbykic--SKA This guy built a rig to run a file against a piece of mild steel with consistent pressure and various techniques (lift vs drag, using it backward, etc) and checked both the amount of material removed and the condition of the files to see if dragging dulled it. TLDR: At least for good-quality machinist files on metal, dragging on the backstroke vs lifting doesn't damage the file.
|
|
#
?
Feb 21, 2022 05:09
|
|
|
tfw when you hear an adult murmur “righty right lefty loosy” when using a screwdriver
|
|
#
?
Feb 21, 2022 05:10
|
|
|
echinopsis posted:tfw when you hear an adult murmur “righty right lefty loosy” when using a screwdriver I use screwdrivers a lot and I still have to think twice when the head of the screw is facing away from me, but I'm also the type of person who physically rotates a paper map so it's facing the same direction I am. Anyway, nothing teaches you the difference between "aluminum tight" and "cast iron tight" quite as effectively as trying to tighten a screw into alumium to cast iron torques. This is also a great way to learn all about heli-coils.
|
|
#
?
Feb 21, 2022 05:19
|
|
|
try to teach kids advanced techniques, like draw filing, or my personal favorite, using two nuts to tighten a length of allthread
|
|
#
?
Feb 21, 2022 07:14
|
|
|
i hadn't heard of draw filing, i'll have to try that some time! they did teach us in high school how to loosen or tighten studs/allthread using two nuts jammed together, that's been super useful over the years
|
|
#
?
Feb 21, 2022 08:55
|
|
|
Corla Plankun posted:jfc between that and the people who don't understand the concept of folders i feel like i am 99% of the way to being that old boomer that thinks everyone around me is incomprehensibly stupid i think there's a relatively small window of say 15-20 years of people who actually understand what a "file" is and how computers organise things. older people think computers are inscrutable boxes and just memorise by rote the six things they have to click to do their job. and younger people think everything just goes into the computer and gets sorted into media libraries that manage themselves
|
|
#
?
Feb 21, 2022 12:45
|
|
|
PIZZA.BAT posted:update: i finally have a minimal tensor set up that appears to be working. i've now also parallelized both the python and java sides of the equation so i'll be able to fully utilize the hardware i have while training instead of everything being choked onto a single core after spending the past month fine tuning the parallelization between the java/python bits, which is necessary because there's a LOT of data moving between the two, i noticed that the best i could push the machine to was ~30% cpu utilization due to the overhead involved in sending messages and all the i/o wait. i was starting to look into ways to speed up this process when a lightbulb suddenly turned on: you only use the python to *train* the model so you can get something you can export in some format. wait a second. these models are probably more broadly adopted by other languages because they have to be used in high-performance production environments. i'm not actually doing any training in the python so i could.... i could probably... god damnit so a few quick google searches later i've found that the generally accepted industry standard is onnx and yes, there's a java library. god damnit. so the past two months of building these message brokers between the two systems, while not being completely worthless, were mostly useless. the only thing i need the python side to do is take the weights and generate an onnx model for the java machine to use. once the java machine has the model all those messages can happen internally which will result in a HUGE performance boost fjeriuwoanvjkflfdnvbjiHGFRUIOPGNFREWUOIka;ruoi; poo poo like this is why i'm convinced no one in the ai space knows what the gently caress they're doing. i've walked several ai people through my architecture and not one of them even hinted that i should move in this direction, but now that i'm here it's extremely obvious that this was the best solution all along. loving 
|
|
#
?
Feb 21, 2022 14:42
|
|
|
echinopsis posted:tfw when you hear an adult murmur “righty right lefty loosy” when using a screwdriver jokes on them, i have 3 different things in my house i intentionally put left-handed threaded fasteners on, so after i die someone's gonna have a hard time
|
|
#
?
Feb 21, 2022 14:54
|
|
|
rather than a field of research, think of ai as a text or spoken input, with the desired output being money from venture capitalists
|
|
#
?
Feb 21, 2022 14:56
|
|
|
GWBBQ posted:rather than a field of research, think of ai as a text or spoken input, with the desired output being money from venture capitalists 100% agreed
|
|
#
?
Feb 21, 2022 15:13
|
|
|
i got everything cut over to being run entirely in the java engine and just finished running some benchmarks. over 5x increase in performance and my cpu was thoroughly pegged at 100% the entire time. gently caress yes. i have an old-rear end 6th gen intel that i'm about to upgrade to a 12th gen by the end of this week too so we'll see how much further it goes with the new hardware. AND THEN i'll start running them on the gpu to see if it goes even faster from there
|
|
#
?
Feb 22, 2022 03:44
|
|
|
man, I’m really jonesing for a vinyl cutter/plotter. those things are hella fun to use and super satisfying to watch as they work
|
|
#
?
Feb 22, 2022 20:32
|
|
|
get yourself an old HP 7475A (RS-232 version) and make some Sharpie holders for it
|
|
#
?
Feb 23, 2022 05:37
|
|
|
im actually looking at a new (or slightly used) graphtec or roland cutter. i dont have the patience to fudge around with stuff not working on the first try anymore
|
|
#
?
Feb 23, 2022 13:06
|
|
|
PIZZA.BAT posted:i got everything cut over to being run entirely in the java engine and just finished running some benchmarks. over 5x increase in performance and my cpu was thoroughly pegged at 100% the entire time. gently caress yes. i have an old-rear end 6th gen intel that i'm about to upgrade to a 12th gen by the end of this week too so we'll see how much further it goes with the new hardware. AND THEN i'll start running them on the gpu to see if it goes even faster from there update: i got another 6x improvement on the new cpu update 2: running it on the gpu made it go slow as dogshit and i'm not really sure why but w/e, the cpu side is fast enough now the real fun begins
|
|
#
?
Mar 3, 2022 05:07
|
|
|
|
| # ? Apr 25, 2024 19:48 |
|
|
New addition to my shop
|
|
#
?
Mar 9, 2022 22:55
|
|
